🚀 go-pugleaf

In article <BEVAN.92Nov11113345@hippo.cs.man.ac.uk>, bevan@cs.man.ac.uk (Stephen J Bevan) writes:
|> In article <BxGBsJ.8qG@unx.sas.com> sasghm@theseus.unx.sas.com (Gary Merrill) writes:
|>    Do you see CS as having any immediate empirical content in the
|>    manner of physics, chemistry, biology, etc.?
|> 
|> What are "etc." in this case?  I'm still waiting to see any mention of
|> a "science" that isn't (IMHO) a "natural science".

Well, there are some who consider the social sciences (anthropology,
sociology, psychology) to be sciences.  Then there is geology, which
has its own problems of methodology (though surely this is a "natural
science" as well).

|> As for empirical content, what about the example of routing via
|> simmulated annealing?  There is no way to check this is any good
|> without experimenting (i.e. writing the code for the router and trying
|> it on as many boards as you can find).  As I noted before you may not
|> consider this CS as a computer was only used to prove the method
|> worked (ignoring the fact that routing has an obvious application in
|> the design of computers).  However, if you do have this view, I would
|> be interested to know under what banner you felt this work lay.

Well, this is exactly the sort of thing I have been fishing for --
but note how long it has taken someone to suggest such an example
rather than purely formal/mathematical methodologies.  (Note also
that it appears to fit a previous suggestion of mine concerning
an analogy between programs and theories.)
-- 
Gary H. Merrill  [Principal Systems Developer, C Compiler Development]
SAS Institute Inc. / SAS Campus Dr. / Cary, NC  27513 / (919) 677-8000
sasghm@theseus.unx.sas.com ... !mcnc!sas!sasghm